Skip to content


ai  101  pytorch  classification  nvidia  cuda  install  tensorrt  yolo  ardupilot  None  ros2  dds  micro ros  xrce  sitl  plugin  SITL  debug  rangefinder  pymavlink  mavros  gazebo  distance sensor  system_time  timesync  cmake  gtest  ctest  cpp  c++  format  fmt  multithreading  spdlog  camera  coordinate system  orb  matching  opencv  build  transformation  computer vision  homography  optical flow  of  trackers  cv  cyclonedds  eprosima  fastdds  simulation  config  ignition  bridge  sdf  tips  ign-transport  sensors  lidar  aptly  apt  encryption  pgp  docker  git  bundle  github  hooks  pre-commit  lxd  container  lxc  x11  profile  vscode  marpit  presentation  marp  markdown  mermaid  video  ffmpeg  gstreamer  cheat-sheet  sdp  v4l2loopback  gi  snippets  cheat Sheet  python  asyncio  future  click  cli  numpy  project  template  black  isort  docs  project document  docstrings  flake8  linter  git-hook  mypy  unittest  pytest  pylint  mock  iterator  generator  logging  tuple  namedtuple  typing  annotation  pyzmq  zmq  msgpack  action  namespace  remap  control2  ros2_control  gdb  qos  tag  plugins  msg  node  zero-copy  shm  tutorial  algorithm  calibration  diff  pid  dev  colcon  colcon_cd  rpi  arm  qemu  settings  behavior  plot  visualization  debugging  diagnostic  diagnostics  tutorials  gst  math  apm  rat_runtime_monitor  web  rosbridge  vue  binding  discovery  gazebo-classic  launch  spawn  cook  gps  imu  ray  gazebo_ros_ray_sensor  ultrsonic  range  ultrasonic  gazebo classic  wrench  effort  odom  ign  gz  xacro  ros_ign  diff_drive  odometry  joint_state  argument  OpaqueFunction  DeclareLaunchArgument  LaunchConfiguration  tmux  nav  slam  test  rclpy  executor  MultiThreadedExecutor  SingleThreadedExecutor  param  dynamic-reconfigure  service  client  setup.py  package.xml  parameter  parameters  custom  msgs  executers  pub  sub  rqt  rviz  rviz2  pose  marker  tf2  deb  package  setup  local_setup  rosdep  package manager  project settings  vcstool  cross-compiler  nano  texture  tmuxp  rootfs  embedded  zah  linux  rm  ubuntu  ip  ss  network  netstat  snap  deploy  ssh  systemd  mkdocs  extensions  socat  networking  serial  udp  tc  mtu  select  px4  robotics  kalman_filter  kalman  filter  control  todo  vscode-ext  json  yaml  schema  yocto  poky  world  gazebo_ros2_control  position_controller  effort_controller  velocity_controller  urdf  gazebo_ros_force  gazebo_ros_joint_state_publisher  robot_state_publisher  joint_state_publisher  projects  vrx  buoyancy 

opencv feature matching

using ORB and Matcher


Hamming Distance#

Hamming Distance measures the similarity between two strings of the same length. Hamming distance is the number of bit positions in which the two bits are different.


BFMatcher#

Brute-Force matching take each feature from image query and compares it to all the other/train image

Matcher result#

match method

DMatch

  • distance
  • imgIdx: train image index
  • queryIdx: query descriptor index
  • trainIdx: train descriptor index

Basic Demo#

import numpy as np
import cv2


# Read the query image as query_img
# and train image This query image
# is what you need to find in train image
# Save it in the same directory
# with the name image.jpg 
query_img = cv2.imread('query.jpg')
train_img = cv2.imread('train.jpg')

# Convert it to grayscale
query_img_bw = cv2.cvtColor(query_img,cv2.COLOR_BGR2GRAY)
train_img_bw = cv2.cvtColor(train_img, cv2.COLOR_BGR2GRAY)

# Initialize the ORB detector algorithm
orb = cv2.ORB_create()

# Now detect the keypoints and compute
# the descriptors for the query image
# and train image
queryKeypoints, queryDescriptors = orb.detectAndCompute(query_img_bw, mask=None)
trainKeypoints, trainDescriptors = orb.detectAndCompute(train_img_bw, mask=None)

# Initialize the Matcher for matching
# the keypoints and then match the
# keypoints
matcher = cv2.BFMatcher()
matches = matcher.match(queryDescriptors,trainDescriptors)

# draw the matches to the final image
# containing both the images the drawMatches()
# function takes both images and keypoints
# and outputs the matched query image with
# its train image
final_img = cv2.drawMatches(query_img, queryKeypoints,
train_img, trainKeypoints, matches[:20],None)

final_img = cv2.resize(final_img, (1000,650))

# Show the final image
cv2.imshow("Matches", final_img)
cv2.waitKey(3000)
#

Reference#